BNP-Response to Acute Heart Failure Treatment Identifies High-Risk Population

Background Using serial measurements of brain natriuretic peptide (BNP) has been proposed as a method to guide therapy for patients treated for acute decompensated heart failure. However, 20-47% of patients do not achieve the target BNP thresholds despite treatment. We hypothesised that BNP unresponsive patients represent a distinct group at high risk for poor outcomes and sought to examine the characteristics and outcomes of this group. Methods In a retrospective study using electronic health record (EHR) data, we examined the outcomes of patients admitted with acute decompensated heart failure. Patients were divided into two groups based on their proBNP response to treatment: (1) pro-BNP responsive to treatment (decrease by at least 30%) and (2) pro-BNP unresponsive to treatment (decrease by less than 30%). The primary outcomes of interest were 180-day mortality and 180-day readmission. Univariate and multivariate Cox proportional hazard models were used to assess the independent association between pro-BNP response to treatment and 180-day mortality and readmission. Adjustment variables included age, gender, Charlson co-morbidity score, admission creatinine, admission haematocrit, ejection fraction, preserved ejection fraction, and LV end-diastolic dimension. Results The total study population included 819 patients with 455 (55.6%) in the pro-BNP responsive group and 364 (44.4%) in the pro-BNP unresponsive group. Admissions whose BNP was unresponsive to treatment had significantly increased risk for 180-day mortality, compared with BNP-responsive admissions (26.4% vs. 13.2%, p < 0.001). Brain natriuretic peptide unresponsiveness remained significantly associated with increased 180-day mortality after adjustment for demographic and clinical characteristics (HRadj =2.19, 95% CI: 1.52-3.14). BNP-unresponsiveness was not associated with significantly increased 180-day readmission rates (HRA(adj) = 1.07, 95% CI: 0.92-1.25). Conclusions Patients whose pro-BNP did not improve by >30% were at increased risk for 180-day mortality, but not 180-day readmission. Thus, BNP-unresponsiveness provides meaningful prognostic information, and it may define a patient population that would benefit from specific therapies to reduce the risk.